JP5072183B2 - Meeting location determination method using spatial semantic modeling - Google Patents

Description

Background of the Invention (1. Field of Invention)
The present invention relates generally to determining meeting locations, and more particularly to methods, apparatus and products for proposing a location of interest to a group of people meeting using a computer application.

(2. Explanation of related technology)
An interesting issue in wireless consumer applications is to suggest places of interest to a group of friends to meet (ie, calculate meeting / meeting places). Various factors must be considered when calculating the meeting place. For example, the current location of all members of this group must be considered. Furthermore, the distance / time of the group members to reach the meeting location and / or the behavioral preferences or trends of each / all members of the group (eg, members prefer outdoor sports such as golf, or , If they prefer to fit in a shopping center) may be beneficial. Furthermore, it is beneficial to converge or tolerate a significant divergence of interest in subgroups.

  Some of these similar considerations apply when suggesting a blind date venue for two people meeting for the first time, or to suggest an action to a single user based on previous behavior profiles.

The prior art method used by the application is for meeting coordinators who examine local yellow pages, some meeting participants, and / or all meeting participants. In this case, each individual selects a favorite place. Thereafter, n squared (n ² ) communications occur between the members of the group to establish consensus and agreed location.

  Such prior art methods have significant limitations. For example, a large amount of data is sent to each user in the group over a low bandwidth network (eg, via phone). In addition, each user must manually search within the yellow page categories and subcategories and select a favorite location. Communication of n squares is required to ultimately draw a difficult agreement for the meeting place. In addition, it is expensive for the user in terms of performance, time, and dollar costs such as wireless communication time and service usage.

(Summary of the Invention)
One or more embodiments of the present invention provide an intelligent list of meeting place proposals that account for a user's current location, individual user's behavioral preferences, and origin / destination distance / progression factors. . No interaction between the users of the group is required other than selecting the final meeting location from the suggested 2-3 locations.

  Two independent axes are used to determine the meeting place: a spatial element and a semantic element. The spatial element determines a central location for use as a seed point for searching for meeting locations of interest. The semantic element calculates a suggested category list given a list of “favorite categories” (representing behavioral preferences) for each member of the group. The proposed category list is a category rank list that represents the convergence of interest in the group.

  By combining semantic and spatial elements, one or more meeting locations can be determined and presented to the user. Using the center position from the spatial element as the seed point, the point of interest belonging to the proposed category from the semantic element is searched. After obtaining a list of potential meeting locations, the locations can be refined based on spatial interaction modeling. In other words, various factors such as meeting place type, meeting place appeal, and distance / progression factors can be used to select or narrow the list of potential meeting places.

(Detailed description of preferred embodiments)
In the following description, reference is made to the accompanying drawings that form a part hereof and in which are shown by way of illustration several embodiments of the invention. Of course, other embodiments can be used and structural changes can be made without departing from the scope of the invention.

(Overview)
One or more embodiments of the present invention calculate an intelligent list of meeting location proposals that account for a user's current location, individual user's behavioral preferences, and origin / destination distance / progression factors. . No interaction between the users of the group is required other than selecting the final meeting location from the suggested 2-3 locations. Such minimal interaction significantly reduces the transfer of data over low-bandwidth wireless networks and saves money for the user. In addition, embodiments increase the price of a commercial establishment that hosts a meeting location through direct personalized marketing. This commercial facility benefits from hosting a group of people known to have a particular interest in its offering. Furthermore, the present invention improves the ease of use of consumer applications by reducing user interface navigation for selecting group behavior (and provides a surprising element).

(Hardware and software environment)
FIG. 1 diagrammatically illustrates a hardware and software environment according to one or more embodiments of the present invention, and in particular, a typical distributed computer system that connects a client computer 104 to a server computer 106 using a network 102. 100 is shown. Network 102 may typically include the Internet, a local area network (LAN), a wide area network (WAN), or the like. The client 104 can typically have a personal computer, workstation, personal digital assistant (PDA), WINCE, PALM device, or mobile phone (eg, a phone that uses wireless application protocol (WAP)). Server 106 may typically include a personal computer, workstation, small computer or mainframe. Moreover, both client 104 and server 106 can receive input (eg, cursor position input) and display a cursor in response to an input device, such as cursor control device 118.

  A network 102 such as the Internet connects a client 104 to a server computer 106. In addition, the network 102 can connect and communicate between the client 104 and the server 106 using radio frequency (RF). Client 104 executes a client application or web browser 108 and can communicate with a server computer 106 running a web server 110. Such a web browser 108 is typically a program such as Netscape Navigator or Microsoft Internet Explorer.

  Web server 110 can host an Active Server Page (ASP) or Internet Server Application Programming Interface (ISAPI) application 112 that can execute scripts. These scripts call objects that execute business logic (also called business objects). The business object then manipulates data in the database 116 via a database management system (DBMS) 114. Alternatively, instead of communicating / obtaining information from the database 116 over the network 102, the database 116 can be part of the client 104 or can be directly connected to the client 104.

  In general, all of these components 108-118 can be a device, medium, signal or carrier, such as a data storage device, a data communication device, a remote computer, or a computer via a network or other data communication device. Have logic and / or data that can be implemented within or retrieved from a coupled device or the like. Further, when this logic and / or data is read, executed and / or interpreted, it results in the steps necessary to implement and / or use the invention to be performed.

  Thus, embodiments of the present invention can be implemented as a method, apparatus, or product that uses standard programming and / or engineering techniques to generate software, firmware, hardware, or any combination thereof. As used herein, the term “product” (or “computer program product”) is intended to include logic and / or data accessible from any computer-readable device, carrier or medium.

  Those skilled in the art will recognize that many variations to this exemplary environment can be configured without departing from the scope of the present invention. For example, using any combination of the above components, or any number of different components, including different logic, data, different peripherals and different devices, as long as they perform similar functions Those skilled in the art will recognize that the present invention can be practiced.

(Software environment)
One or more embodiments of the present invention provide that an application (eg, application 112 on server 106 or application 108 on client 104) meets (eg, two to four) meeting places (eg, a group of users, or It makes it possible to propose a short list of two or more users' blind date meeting places. Such an application 108 or 112 is provided with a list representing a group of users (eg, client 104), the user's current location, and the favorite categories / trends of individual users. Alternatively, given the single user's current location and behavior profile, the application 108 or 112 can suggest individual user behavior.

Two independent axes can be used to determine a list of suitable meeting places: a spatial element and a semantic element.
Spatial Elements Spatial elements of the present invention are used to determine the central location to use as a seed point for searching for meeting locations of interest. n user positions {(x1, y1), (x2, y2),. . . . If (xn, yn)} is given, the spatial element has a central position.

  The center position used as a seed point can be calculated using a variety of different methods. Each method uses the seed point for the meeting location search and does not itself represent the meeting position. Two examples of such methods include centroid calculation and median calculation.

  In the calculation of the center of gravity, the average / intermediate of the position coordinates of the user is taken. Therefore, X center = intermediate (x1, x2,... Xn) and Y center = intermediate (y1, y2,... Yn). By taking the average / intermediate for each coordinate point for various user positions, the centroid between all positions can be determined. The center of gravity does not represent the actual meeting place; in practice, the center of gravity is the climate (such as in a cave, the center of a lake / ocean, on a mountain, in Antarctica, etc.) / It can be inaccessible due to geology.

  FIG. 2 is a flowchart illustrating centroid calculation according to one or more embodiments of the present invention. In step 200, (X, Y) coordinates for various user positions are obtained. In step 202, the average / intermediate for the X coordinate is obtained. In step 204, an average / intermediate for the Y coordinate is obtained. In step 206, the center position is set to the obtained (X, Y) value (ie, (XAVG, YAVG)).

  In the median calculation, the user's position coordinates are sorted in ascending / descending order, and the median of each coordinate list is used as the center position. FIG. 3 is a flowchart illustrating median calculation according to one or more embodiments of the invention. In step 300, the (X, Y) coordinates for each user's position are obtained. In step 302, the (eg, X and Y) coordinates are sorted independently in ascending or descending order. In step 304, the median value of each list is determined. In step 306, the center position is set to the median value.

  The description of the center point calculation described above uses (X, Y) coordinates. However, other means of describing the geographic location (eg latitude and longitude, 3D coordinates (X, Y, Z), etc.) also described calculations and methods (or other methods of determining the center location) Can be used.

  Sophisticated methods for determining the center location use street networks (eg, sign speed in the street network), (current or predicted) traffic times and / or (current or predicted) weather conditions. be able to. In such an embodiment, the center position can be adjusted to reflect to each user a position that takes approximately the same time to reach. For example, if the street network forces a person to travel 50 miles off the road to reach a position of only 10 miles from the center of gravity (determined above), embodiments of the present invention would be Center location / center of gravity can be adjusted to account for unusual street networks and transit times. Similarly, when using aircraft operation, the center position can also be adjusted in this case by using the position from each party to the plane and the operation time from each airport to the arrival city.

  In addition, the center location can be adjusted using prices and costs associated with arrival at the destination. For example, the center location can be adjusted using airplane / train operating costs so that the pricing of each person arriving at the center location remains relatively uniform.

Therefore, various elements and methods can be used to accurately determine the center location used as the seed point and reflect this in determining the meeting location.
Semantic Elements The semantic elements of the present invention are used to form a proposed category list. The proposed category list is a rank list of categories that represents the convergence of interest for a group (or a major subgroup if there is a significant divergence of interest within the group). (Also referred to as a “favorite category”) (ie representing behavior preferences for each member of the group, eg representing Chinese restaurants, cinemas, golf, parks, shopping centers, tennis, etc.) The semantic element calculates a rank list of categories or behavior preferences.

  Semantic elements can be calculated using various methods. Hierarchical trees have historically proven to be a good example of acquiring semantic relationships in other fields (eg, taxonomic classification of organisms in ecology). Hierarchical graphs / trees allow you to see the relationships and show them at the appropriate level of specialization (for example, higher order nodes that give an overall view and lower order nodes that give an increased level of specialization) Can do. Therefore, semantic search can be aided using a hierarchical graph / tree. In this regard, a hierarchical graph can be used to obtain an overall semantic relationship between behavioral preferences and to identify a group (eg, pointed convergence of interest). For example, a grouping along a subtree can represent a convergence of interest (similar to a type in taxonomic classification). Also, neighboring nodes can inform additional semantic distance information. For example, indoor entertainment behavior nodes (eg, movie theaters, sports bars, etc.) can be set adjacent to each other and separated from outdoor entertainment behavior nodes (eg, golf, tennis, etc.).

  To use a hierarchical graph, points of interest are collected from various users and classified into categories. Interest points and favorite categories can be retrieved or used from a user list of favorite locations identified on a web browser or a WAP-enabled phone mini-browser. Therefore, the user can record / store and use his / her browsing / use history.

  Alternatively, points of interest can be obtained from the user via questions from the server 106 or from software on the client 104. For example, due to the limited memory and screen space available on WAP-enabled phones, it is possible to collect behavioral preferences (eg, a list of menus displayed on a continuous screen sent to the client 104) You can ask multiple questions from overview to details. In particular, the user can be asked to select a category of interest (e.g., given a selection set of restaurants, retail stores, entertainment, etc.). When the user answers, an additional list of subcategories can be presented to the user for selection. The selection can be continuously narrowed until a list of actual jobs / locations is presented to the user (eg, from a white page or yellow page directory). After the user selects the actual location, the map, direction, reservation, etc. can be presented to the user.

  Once the category list is created and user preferences are incorporated into the category list / hierarchy graph, the categories can be weighted based on the number of users who prefer each category. FIG. 4 shows an example of a result obtained by conducting a survey with a friend group regarding the friend category. 42 different categories were obtained and classified. For example, all preferences were obtained from a plurality of users and arranged in the hierarchical graph 400 based on interview surveys. As shown, the points of interest (POI) were categorized at the first level as a Restaurant 402, a Shop 404, a Bussiness 406, and an Entertainment 408. As needed, the categories were further subdivided into sub-categories (eg, Asian Restaurant 410 and American 412 Restaurant). The suitability of further classification into subcategories can be based on a variety of factors including the user's background, a list of favorites, and a specified classification level. Alternatively, the number of subcategories can be based on questions and answers given to / by the user. In yet another embodiment, any number of subcategories can be configured.

  Next, the input given by the user's favorite profile is overwritten on the hierarchical graph. In this case, the categories were weighted based on the number of users who liked each category. For example, 5 users prefer Chinese restaurants and 11 users prefer cinemas.

FIG. 5 is a flowchart illustrating the use / creation of a hierarchical graph used in a semantic element in accordance with one or more embodiments of the present invention. In step 500, the points of interest are classified into categories. In step 502, if appropriate, the points of interest are further classified into subcategories. In step 504, the behavioral preference from the user is overwritten on the hierarchical graph / category list. By overriding behavior preferences, each category can also be weighted based on the number of users who prefer each category. The weighted category list can also be referred to as a behavior preference rank list. Thus, this rank list represents a convergence of interest for multiple (ie, two or more) users.
Meeting place determination Meeting places (or a list of meeting places) can be determined / presented by combining spatial and semantic elements. Using the center position as a seed point, the application searches for one or more points of interest belonging to the proposed category list. For example, if the central location is Powell and Market in San Francisco and the first proposed category is a sports bar, the meeting location searches for the sports bar closest to Powell and Market.

  After obtaining a center location and a rank list of behavioral preferences (eg, a weighted hierarchy graph), the rank list is examined in detail to determine one or more categories for potential meeting locations. To find the appropriate category, search using a hierarchical tree (eg, tree 400 in FIG. 4) to find the heaviest subtree (or trail). Next, one or more categories for the meeting location can be selected by setting the fulcrum on the heaviest trail or subtree.

  This fulcrum is set so as to form an equilibrium between categories that are too specialized (for example, POI.restaurant.Asian style.Chinese style.high-ranking bureaucracy) and categories that are too general (POI.restaurant). A threshold or minimum / maximum satisfaction level can be used to form such an equilibrium. The threshold level can reflect the number of people that need to be satisfied. In this regard, subcategories can be selected sequentially as long as the result of decomposing and selecting the subcategories does not result in user satisfaction below a certain percentage. For example, if a threshold value of 30% is selected, the selected categories are continuously decomposed until the result of the selection of subcategories results in the selection of categories that satisfy less than 30% of users. In this regard, as the threshold is set to a decreasing value, fewer people / users are more satisfied with the selected category. The threshold level can be set arbitrarily or can be adjusted for each user of the system.

  In FIG. 4, by examining the tree in detail and setting fulcrums, two categories for meeting places, namely POI. Restaurant. Asian style. Chinese style (POI. Restaurant. Asian. Chinese) 414 and POI. Entertainment. Movie theater (POI.Entertainment.Cinema) 416 was obtained. After multiple categories are selected, the user can be given the option to select a favorite category. In addition (or alternatively), the meeting location can be determined by searching for a specific meeting location within the specified category closest to the central point. In FIG. 4, the meeting place is determined by searching for a movie theater or a Chinese restaurant closest to the central position (−122.511, 37.751).

  FIG. 6 is a flowchart illustrating meeting location determination according to one or more embodiments of the present invention. In step 600, the heaviest subtree (or trail) is found. In step 602, a fulcrum is set for the found subtree (or trail). As described above, setting a fulcrum includes selecting one or more categories that satisfy a minimum threshold percentage of users, where the selection of subcategories within the selected category is minimal. Satisfy users below the threshold percentage. Step 604 selects a meeting location from the category identified by the fulcrum closest to the center position.

FIG. 7 is a flowchart illustrating an overview of the steps performed to determine a meeting location according to one or more embodiments of the present invention. In step 700, the position of each of two or more people is acquired. In step 702, a center position is determined based on the acquired position. In step 704, a list of behavioral preferences for two or more people is obtained. In step 706, a rank list of behavior preferences is calculated based on the list obtained from the user. A rank list represents a convergence of interest for two or more people. In step 708, one or more meeting locations are determined by combining the center location and the rank list.
Refinement Based on Spatial Interaction Modeling After calculating more than one meeting location, further refinements can be applied to determine which particular meeting location should be selected. For example, with further refinement, a particular Chinese restaurant can be selected from three possible candidates available near the meeting place. Such a refinement can be determined using spatial interaction modeling.

  Spatial interaction modeling can take into account various elements / attributes. Such factors include trends in the flow characteristics of the origin / meeting location, attractiveness of the origin / meeting location, and / or general separation / distance (or progression) impediments to the origin / meeting location from the destination. Can do.

  When considering the flow characteristics or trends of the origin / meeting location, the user's overall profile can be examined. In this regard, the user can select a specific profile for the meeting. Alternatively, the profile can be automatically determined based on the background of users in the group, the time / date planned for the meeting, and the like. For example, if a group of users is a business visitor group with a spending budget, or a teenage group that expects a fun time but is limited by spending money, or a nice budget to go out on a great evening. Decide and consider whether or not you are expecting an adult couple. Based on these considerations, a specific meeting place can be selected that is superior to another meeting place.

  When considering the attractiveness of a meeting place, various attributes of the meeting place can be examined and compared with user expectations. For example, the floor area of a restaurant, the waiting time of a restaurant, or the level of service provided by a store owner can be compared to what is expected by a group of users. These expectations can be based on information selected / given directly by the user. Alternatively, the user's expectations can be determined based on a previous history of places visited by the user (e.g., restaurants that have eaten). The career can be further analyzed and compared to available guides that provide details about a particular location (eg, restaurant guides available from the Zagat® Survey).

  Additional considerations that can narrow or select the final meeting location can evaluate general spatial separation of the meeting location or distance / progression factors from the user location. For example, if one user can drive to the location, while another user may be forced to fly by plane, the cost difference can be taken into account. Alternatively, if the distance from the user location to the central location requires a plane trip for all (or more) parties, the embodiment will force the meeting location to force only one user to fly. Can be adjusted. In another example, obstructive factors such as weather conditions may force one participant to wait for the storm to stop before traveling, while another participant may It may be starting a trip to the participants. The meeting place can be adjusted in such a situation.

Any type and number of factors and circumstances can be used to narrow the meeting place. Therefore, the present invention is not limited to the above-described narrowing down.
Conclusion This concludes the description of the preferred embodiment of the present invention. The following describes some other embodiments that accomplish the present invention. For example, any type of computer such as a mainframe, small computer or personal computer, or a computer configuration such as a time-sharing mainframe, local area network or stand-alone personal computer can be used in the present invention. In summary, embodiments of the present invention provide a method for determining a meeting location based on a list of behavioral preferences and locations for multiple users.

  The foregoing description of preferred embodiments of the present invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in light of the above teaching. It is intended that the scope of the invention be limited not by this detailed description, but rather by the appended claims.

  Reference is now made to the drawings. In the figure, the same reference numerals represent corresponding parts.

FIG. 1 diagrammatically illustrates a hardware and software environment according to one or more embodiments of the present invention. FIG. 2 is a flowchart illustrating centroid calculation according to one or more embodiments of the present invention. FIG. 3 is a flowchart illustrating median calculation according to one or more embodiments of the invention. FIG. 4 illustrates an example of the results obtained by interviewing a friend group for a friend category in accordance with one or more embodiments of the present invention. FIG. 5 is a flowchart illustrating the use / creation of a hierarchical graph used in a semantic element in accordance with one or more embodiments of the present invention. FIG. 6 is a flowchart illustrating meeting location determination according to one or more embodiments of the present invention. FIG. 7 is a flowchart illustrating an overview of the steps performed to determine a meeting location according to one or more embodiments of the present invention.

Claims (15)

A method for determining a meeting place, said method being executed by a computer, said computer comprising input means and processor means,
The method
The input means retrieves geographical location data of each of the two or more client devices;
The processor means calculates geographic center location data based on the retrieved geographic location data of each of the two or more client devices;
The processor means sends a continuous behavior preference option to each of the two or more client devices; each of the two or more client devices responds with a behavior preference for each option; Retrieving a behavior preference data list for each of the two or more client devices by placing each behavior preference received from each of the two or more client devices in a respective data list;
The processor means calculating a ranked data list of behavioral preferences based on the retrieved behavioral preference data list, the computing being most frequently occurring in the retrieved data list; Rank the behavioral preference to the highest rank, rank the behavioral preference that appears next highest to the next highest ranking, and rank the behavioral preference that appears least frequently in the retrieved data list to the lowest rank Continue to rank sequentially until
The processor means calculates a distance between a geographical center position represented by the calculated geographical center position data and a geographical position associated with a behavior preference ranked in the ranked data list; Calculating one or more geographic meeting locations , wherein the calculated one or more meeting locations are ranked in the highest rank closest to the calculated geographic center location A method comprising:   The method of claim 1, wherein calculating the geographic center location data includes calculating geographic average coordinates for the retrieved geographic location data. Calculating the geographic center location data comprises:
Generating a sorted geographic coordinate data list by independently sorting the geographic coordinates of the retrieved geographic location data;
The method of claim 1, comprising: determining a geographic median value for each of the sorted geographic coordinate data lists as the geographic center location.   The method of claim 3, wherein the geographic coordinates are sorted in ascending order.   The method of claim 3, wherein the geographic coordinates are sorted in descending order.   The method of claim 1, wherein the geographic center location data is based on sign speed in a street network.   The method of claim 1, wherein the geographic center location data is based on current transit time.   The method of claim 1, wherein the geographic center location data is based on weather conditions.   The method of claim 1, wherein calculating the ranked data list of behavioral preferences obtains an overall semantic relationship between behavioral preferences by using a hierarchical graph.   The computing the ranked data list of behavior preferences includes classifying points of interest into categories from the behavior preferences, wherein the categories are weighted based on the number of users who prefer each category. The method described in 1. Determining the one or more geographic meeting locations is
Selecting one or more categories with the greatest weight;
The method of claim 10, comprising: calculating one or more geographic meeting locations closest to the geographic center location from the selected category.   Selecting the one or more categories having the maximum weight includes selecting one or more categories that satisfy a minimum threshold percentage of the client device, the one or more selected The method of claim 11, wherein selection of subcategories in a selected category satisfies less than the minimum threshold percentage of the client device.   The method of claim 1, wherein the one or more geographic meeting locations are calculated based on the attraction of the geographic meeting location.   The method of claim 1, wherein the one or more geographic meeting locations are determined based on a distance obstruction factor to the one or more geographic meeting locations for one or more client devices.   The method of claim 1, wherein the one or more geographic meeting locations are determined based on the type of meeting to be performed.

Images